Physical neural network liquid state machine utilizing nanotechnology
AssigneeKnowmTech, LLC
InventorAlex Nugent
Life of this patent
Estimating the term…
Abstract
A physical neural network is disclosed, which comprises a liquid state machine. The physical neural network is configured from molecular connections located within a dielectric solvent between pre-synaptic and post-synaptic electrodes thereof, such that the molecular connections are strengthened or weakened according to an application of an electric field or a frequency thereof to provide physical neural network connections thereof. A supervised learning mechanism is associated with the liquid state machine, whereby connections strengths of the molecular connections are determined by pre-synaptic and post-synaptic activity respectively associated with the pre-synaptic and post-synaptic electrodes, wherein the liquid state machine comprises a dynamic fading memory mechanism.
Claims19
An electromechanical liquid state machine utilizing nanotechnology, comprising:
- at least one pre-synaptic electrode and at least one post-synaptic electrode having a connection gap therebetween;
- a liquid dielectric solution comprising a mixture of a plurality of nanoconductors and a liquid dielectric solvent, wherein said plurality of nanoconductors are located in and free to move about within said liquid dielectric solution, said liquid dielectric solution disposed in said connection gap between said at least one pre-synaptic electrode and said at least one post-synaptic electrode;
- and
- a mechanism for applying an electric field across said connection gap, said mechanism electrically connected to said at least one pre-synaptic electrode and said at least one post-synaptic electrode, whereby said electric field induces a dipole in each nanoconductor among said plurality of nanoconductors only when said plurality of nanoconductors is located within said liquid dielectric solution, thereby aligning said plurality of nanoconductors within said liquid dielectric solution and attracting said plurality of nanoconductors to said connection gap in order to provide to neural network nanoconnections of a connection network between said at least one pre-synaptic electrode and said at least one post-synaptic electrode within said liquid dielectric solution, said connection network, said liquid dielectric solution, said plurality of nanoconductors, said at least one pre-synaptic electrode and said at least one post-synaptic electrodes electromechanically operable in combination with one another to comprise said electromechanical-based liquid state machine, wherein at least one neural circuit thereof comprises a plurality of neural network connections and at least one perceptron adjusting at least one synaptic weight to produce a predetermined output, which stores via patterns of neural activations of said neural network nanoconnections, a recent past history of said electromechanical-based liquid state machine.
An electromechanical liquid state machine based on nanotechnology, comprising:
- a liquid dielectric solution composed of a plurality of nanoconductors and a liquid dielectric solvent;
- at least one pre-synaptic electrode and at least one post-synaptic electrode having a connection gap therebetween, wherein said liquid dielectric solution is disposed in said connection gap;
- a mechanism for providing a dielectrophoretic force that permits nanoconductors among said plurality of nanoconductors align to form neural network nanoconnections between said pre-synaptic and post-synaptic electrodes within said liquid dielectric solution thereby providing a liquid state machine that stores via patterns of neural activations, a recent past history of said liquid state machine;
- and
- a plurality of perceptrons in communication with said liquid state machine and said physical neural network, wherein said plurality of perceptrons extracts at least one state of said liquid state machine from said liquid state machine.
A liquid state machine based on nanotechnology comprising:
- a physical neural network comprising a liquid dielectric solution composed of a plurality of nanoconductors and a liquid dielectric solvent, said liquid dielectric solution disposed in a connection gap formed between at least one pre-synaptic electrode and at least one post-synaptic electrode;
- a mechanism for providing a dielectrophoretic force for aligning said nanoconductors to form random neural network nanoconnections of a connection network between said pre-synaptic and post-synaptic electrodes within said liquid dielectric solution when an electric field is applied across said connection gap, such that said random neural network nanoconnections are strengthened or weakened according to said application of said electric field, a frequency of said electric or a combination thereof to provide said random neural network nanoconnections;
- a gate located adjacent said connection gap in association with an insulator located between said gate and said connection gap, a logic circuit located to said gate, and wherein said at least one pre-synaptic electrode comprises a source and said at least one postsynaptic electrode comprises a drain, wherein said liquid state machine stores via patterns of neural activations of said physical neural network, a recent past history of said liquid state machine.
Dependencies are inferred from the claim text.